An Integrated CT-based Image-Guided Neurosurgical System In this SBIR Phase IIb proposal Xoran intends to commercialize a compact and affordable, yet highly- functional, system to provide real time image updates and navigation guidance in support of minimally invasive cranial and spinal neurosurgical procedures. The effort builds on previously developed compact and portable flat-panel Computed Tomography (CT) technology which has been commercialized for hard tissue applications, and incorporates work done in earlier phases of this project to generate viable high- quality images of the soft tissue structures in the brain. Intraoperatively obtained images tightly integrated into an onboard surgical navigation will provide updated instrument localization using next generation electromagnetic tool tip guidance. Workflow optimizations become possible when the imaging and guidance are one device, including fast local image updates, automatic image-to-world registration, as well as speed and simplicity of use. The project includes expansion of the system capabilities to facilitate precise minimally-invasive surgical removal of tumors in both the head and spine. It incorporates a machine-learning based deep neural network method for image finalization to allow high quality, low radiation image updates. The three-year project involves meeting technical milestones of system development including imaging capability, registration, navigation accuracy, speed, workflow, radiation dose considerations and cost. Clinical evaluations will take place at University of Michigan, and a team of consulting physicians has been assembled for oversight, input and feedback.